Fusion Gene Detection Enhances Clinical Outcome Prediction of Hepatocellular Carcinoma Through Machine Learning...

Area of Science:

• Oncology
• Genetics
• Machine Learning

Background:

• Hepatocellular carcinoma (HCC) is a highly lethal malignancy with challenging outcome assessment.
• Predicting recurrence and survival rates in HCC patients undergoing surgery is crucial for effective treatment planning.

Purpose of the Study:

• To evaluate the predictive power of a panel of 20 fusion genes for HCC recurrence and survival.
• To assess the efficacy of machine learning models in predicting HCC outcomes.
• To determine if combining fusion gene data with clinical parameters enhances predictive accuracy.

Main Methods:

• Analysis of 20 fusion genes in 200 HCC samples.
• Application of machine learning models to predict recurrence and survival rates.
• Comparison of fusion gene panel predictions with established clinical parameters (Milan criteria, serum AFP, pathology grade).

Main Results:

• Fusion genes, Milan criteria, serum AFP, and pathology grade showed moderate predictive accuracy for HCC recurrence.
• Combining fusion genes with clinical parameters significantly improved prediction accuracy.
• Fusion gene models predicted 3-year HCC survival with 72.4% accuracy, outperforming Milan criteria, pathology grade, and serum AFP alone.
• Combined models demonstrated superior predictive performance compared to individual parameters.

Conclusions:

• Fusion gene status holds promise as a novel biomarker for assessing HCC clinical outcomes.
• Integrating fusion gene analysis with clinical parameters offers a more accurate approach to predicting HCC recurrence and survival.
• This approach may lead to improved patient management and treatment strategies for hepatocellular carcinoma.